Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. If you'd prefer to explore, try our test area to get started. And see here for current known issues.

New Member

Active PIM / PGAGENT on PG side B swich over to PG side A whenever the side A server is rebooted

Hi Team,

 

I have a weird behaviour on a newly installed Peripheral Gateway VRU in duplex mode duplex version 8.0.

Whenever the PIM and the PGAGENT is active on side B (side A is Idle) and I restart the side A Server, the MDS process on side B is terminated when the side A finish booting.This behaviour is only found when the side A server is rebooted.

Please note that on both side there is no preferred side configured on PG side A and B.

Underneath is the PGAGENT's trace from side B perspective. I have also enclosed the MDS and OPC trace... Has anybody found this kind of behaviour or got any idea of what is going on ?

Thank you !

 

Regards

 

Nick

 

-------------------------------------------------------------------------------------

Events from April 10, 2014:

14:00:43:312 PG3B-pgag Trace: TOS requested but no TOS connection exists.  Assuming unreachable. 
14:00:43:312 PG3B-pgag Trace: TOS requested but no TOS connection exists.  Assuming unreachable. 
14:00:43:312 PG3B-pgag DMP Agent process activated. 
14:00:43:328 PG3B-pgag Connection to Central Controller side B failed (high priority). 
  Last EMT Error [64]: The specified network name is no longer available. 
14:00:43:328 PG3B-pgag Connection to Central Controller side A failed (medium priority). 
  Last EMT Error [64]: The specified network name is no longer available. 
14:00:43:328 PG3B-pgag Connection to Central Controller side B failed (low priority). 
  Last EMT Error [64]: The specified network name is no longer available. 
14:00:43:328 PG3B-pgag Connection to Central Controller side A failed (low priority). 
  Last EMT Error [10053]: An established connection was aborted by the software in your host machine. 
14:00:43:328 PG3B-pgag Connection to Central Controller side B failed (medium priority). 
  Last EMT Error [64]: The specified network name is no longer available. 
14:00:43:328 PG3B-pgag Connection to Central Controller side A failed (high priority). 
  Last EMT Error [64]: The specified network name is no longer available. 
14:00:53:593 PG3B-pgag Connection to Central Controller side B established (medium priority). 
14:00:53:593 PG3B-pgag Connection to Central Controller side B established (high priority). 
14:00:53:593 PG3B-pgag Connection to Central Controller side B established (low priority). 
14:00:53:593 PG3B-pgag Connection to Central Controller side A established (medium priority). 
14:00:53:593 PG3B-pgag Connection to Central Controller side A established (high priority). 
14:00:53:593 PG3B-pgag Connection to Central Controller side A established (low priority). 
14:00:53:609 PG3B-pgag Path to Central Controller side B established. 
14:00:53:625 PG3B-pgag Path to Central Controller side A established. 
14:00:53:625 PG3B-pgag Central Controller side B is in service. 
14:00:53:625 PG3B-pgag Central Controller side B reports good response time. 
14:00:53:625 PG3B-pgag Initializing message stream with side B of the Central Controller. 
14:00:53:625 PG3B-pgag Central Controller side A is in service. 
14:00:53:625 PG3B-pgag Central Controller side A reports good response time. 
14:00:53:718 PG3B-pgag Path to Central Controller side A is now idle. 
14:00:53:718 PG3B-pgag Path to Central Controller side B is now active. 
14:04:39:718 PG3B-pgag Connection to MDS process has failed. 
  Last EMT Error [-519897069]: The remote process has exited. 
14:04:39:718 PG3B-pgag MDS is out of service. 

14:04:39:718 PG3B-pgag DMP Agent process being deactivated. 
14:04:39:718 PG3B-pgag Path to Central Controller side A disconnected. 
14:04:39:718 PG3B-pgag Path to Central Controller side B disconnected. 
14:04:39:718 PG3B-pgag Message stream to Central Controller has been broken. 
14:04:49:984 PG3B-pgag Connection to Central Controller side A established (high priority). 
14:04:49:984 PG3B-pgag Connection to Central Controller side A established (medium priority). 
14:04:49:984 PG3B-pgag Connection to Central Controller side A established (low priority). 
14:04:49:984 PG3B-pgag Connection to Central Controller side B established (medium priority). 
14:04:49:984 PG3B-pgag Connection to Central Controller side B established (low priority). 
14:04:49:984 PG3B-pgag Connection to Central Controller side B established (high priority). 
14:04:55:718 PG3B-pgag Connection to MDS process established. 
14:04:55:718 PG3B-pgag Reconnection to MDS process completed successfully. 
14:04:56:359 PG3B-pgag MDS is in service. 

 

 

Everyone's tags (1)
3 REPLIES
Silver

14:00:43:312 PG3B-mds Trace:

14:00:43:312 PG3B-mds Trace: Synchronizer breaking dedicated path. 
14:00:43:312 PG3B-mds Connectivity with duplexed partner has been lost due to a failure of 
the private network, or duplexed partner is out of service. Last EMT Error [64]: The specified
network name is no longer available. 
14:00:43:312 PG3B-mds Trace: Synch high priority transmit thread exiting. 
14:00:43:312 PG3B-mds Trace: Synch low priority transmit thread exiting. 
14:00:43:312 PG3B-mds Trace: Synch changing to testing state. 
14:00:43:312 PG3B-mds Trace: Synch medium priority transmit thread exiting. 
14:00:43:312 PG3B-mds Trace: 0 - messages kept in local ordering queue. 
14:00:43:312 PG3B-mds Trace: 2 - messages cleared from peer ordering queue. 
14:00:43:312 PG3B-mds Trace: 0 - messages cleared from timed delivery queue. 
14:00:43:312 PG3B-mds Initiating test of peer Synchronizer. 
14:00:43:312 PG3B-mds Trace: Sending TOS request: sequence = 1. 
14:00:43:312 PG3B-mds Trace: Received TOS response: sequence=1 status=UNREACHABLE. 
14:00:43:312 PG3B-mds Trace: Sending TOS request: sequence = 2. 
14:00:43:312 PG3B-mds Trace: Received TOS response: sequence=2 status=UNREACHABLE. 
14:00:43:312 PG3B-mds Peer Synchronizer was found to be unreachable. 
14:00:43:312 PG3B-mds Trace: Synch changing to isolated-enabled state. 
14:00:43:312 PG3B-mds Synchronizer switching to non-duplex operation. 

....

14:02:40:109 PG3B-mds Trace: Ignoring 'SetSystemTimeAdjustment', Time Synchronization is Disabled 
14:02:40:109 PG3B-mds Trace: Synchronizer is releasing clock master mutex. 
14:02:50:109 PG3B-mds Trace: Ignoring 'SetSystemTimeAdjustment', Time Synchronization is Disabled 
14:02:50:109 PG3B-mds Trace: Synchronizer has obtained the clock master mutex. 
14:02:55:093 PG3B-mds Trace: Node Manager ping received. 
14:02:55:109 PG3B-mds Trace: Ignoring clock master status event in state Isolated-Enabled 

If I'm understanding your post correctly, everything seems normal and working as designed. What you're seeing is the peer-to-peer redundancy that the message delivery system (MDS) offers; MDS works on the private network. What you have is side B of the PGAG process active, so the test other side (TOS) gets initiated over the public network when side A reboots and goes from duplex to non-duplex mode. Additionally, you'll see that you go from In-Service Paired to Isolated, and again, this is by design. Once side A comes back up, the enabled role should go back to side A IF MDS side B was the enabled side. 

New Member

Hi Omar,Thank you for your

Hi Omar,

Thank you for your answer ! 

In my lab, I tried to duplicate this behaviour, and whenever the PG server Side A is restarted, and the PIM and the PGagent become active on Side B, the PIM andthe  PGagent never switch back to Side A once the reboot is over.

This is what bother me, and I couldn't find any official statement from Cisco saying that Side A is always the preferred side...

 

Now looking back to the PGagent trace, the MDS process side B dies whenever the PG Side A is back to normal :

14:04:39:718 PG3B-pgag Connection to MDS process has failed. 

  Last EMT Error [-519897069]: The remote process has exited. 
14:04:39:718 PG3B-pgag MDS is out of service. 

I don't know if this is a normal behaviour...

 

Regards

Nick

Silver

Documentation attached

Documentation attached

1022
Views
0
Helpful
3
Replies